The proposed research develops computer methodology for application to problems in the neurosciences. The areas of computer methodology include: data acquisition, signal analysis, interactive graphics, and cooperative computer systems. Applications are in support of research in basic neurophysiology and clinical neurology. Analytical and computational techniques suitable for the study of both multivariable continuous and point-process physiological data will be developed to study the relationships among spike trains of discharging neurons; long-term periodicities in autonomic and central nervous system variables in sleep; interrelationships among physiological variables across sleep states; evoked responses in the human EEG, particularly in vestibular and somatosensory systems. Further development of software for a cooperative computer system, consisting of a laboratory computer, an interactive graphics terminal, and a large time-shared computer is proposed. The use of interactive graphics as a tool for modeling, simulation, analysis, and information retrieval will be pursued. This system will be used to develop a computer-graphics method for displaying brain cross-sections, for use during human stereotaxic surgery. It will also be used for testing neuronal models against experimental data, using modeling, simulation and identification techniques. The application of microprocessors for the acquisition and analysis of neurophysiological signals will be investigated.